AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

FIBER: A Differentially Private Optimizer with Filter-Aware Innovation Bias Correction

Duc Dm, Thao Do, Minh Son Hoang, Anh Le Duc Tran, Daeyoung Kim, Huy Nguyen

ICML 2026 · Federated learning and privacy › Differential privacy · approach: DP-SGD refinements · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
innovation filtering (FiBeR) instead of DiSK gradient-state smoothing (ablation)DP-AdamW · CIFAR-10 CNN5 eps=0.5accuracy (%): 59.70 -> 65.32Improved
innovation filtering instead of DiSK smoothing (ablation)DP-AdamW · CIFAR-10 CNN5 eps=2accuracy (%): 68.80 -> 71.39Improved
innovation filtering instead of DiSK smoothing (ablation)DP-AdamW · CIFAR-10 CNN5 eps=8accuracy (%): 74.90 -> 73.19Hurt
raise innovation gain omega from 0.60 to 0.90 (ablation)FiBeR · CIFAR-10 CNN5 eps=1accuracy (%): 68.86 -> 70.76Improved
raise innovation gain omega from 0.90 to 0.99 (ablation)FiBeR · CIFAR-10 CNN5 eps=1accuracy (%): 70.76 -> 70.62No clear effect
add filter-aware second-moment correction (ablation)FiBeR · CIFAR-10 across privacy budgetsaccuracyImproved
add FiBeR on top of DiSKDiSK · ViT-small CIFAR-100 fine-tuning eps=0.5accuracy (%): 83.5 -> 88.6Improved
add FiBeR on top of DiSKDiSK · ViT-small CIFAR-100 fine-tuning eps=8accuracy (%): 88.5 -> 90.5Improved
add FiBeRDiSK · ImageNet-1k ViT-smallaccuracy: 39 -> 44Improved
add FiBeRDPAdamW · GLUE QNLI eps=1accuracy: 86.0 -> 90.5Improved
add FiBeR vs DiSKDiSK · GLUE MNLI eps=6.7accuracy: 84.8 -> 84.7Hurt
add FiBeR vs DiSKDiSK · E2E BLEU eps=3BLEU: 68.35 -> 67.57Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy